Opinion
Case No. 5:13-cv-00825-PSG
01-23-2015
SENTIUS INTERNATIONAL, LLC, Plaintiff, v. MICROSOFT CORPORATION, Defendant.
ORDER DENYING MOTION TO EXCLUDE
(Re: Docket No. 130)
Long a staple of trademark, false advertising and antitrust cases, consumer surveys are now de rigueur in patent cases as well. Unlike some of the more esoteric tools used in such cases, surveys are not exactly unusual or unfamiliar to the layperson. Starbucks sends its customers surveys inquiring about their experiences with baristas. Pollsters call voters asking about their opinions on local politicians. And 60 Minutes and Vanity Fair conduct polls asking people about everything from whether we will discovers aliens before we discover a cure for cancer to whether celebrities are being greedy or smart when they sell their wedding photographs. So it is hardly surprising that the Ninth Circuit has concluded that "[u]nlike novel scientific theories, a jury should be able to determine whether asserted technical deficiencies undermine a survey's probative value."
See "The 60 Minutes/Vanity Fair Poll," Vanity Fair, Nov. 1, 2009, http://www.vanityfair.com/magazine/2009/12/60-minutes-poll-200912; "Want, Want, Want," Vanity Fair, Dec. 2014, http://www.vanityfair.com/magazine/2014/12/greed-60-minutes-poll_slideshow_item3_4.
Southland Sod Farms v. Stover Seed Co., 108 F.3d 1134, 1143 n.8 (9th Cir. 1997); see also ResQNet.com, Inc. v. Lansa, Inc., 594 F.3d 860, 869 (Fed. Cir. 2010) (noting that "[d]etermining a fair and reasonable royalty is often ... a difficult judicial chore, seeming often to involve more the talents of a conjurer than those of a judge").
On behalf of Plaintiff Sentius International, LLC, William Wecker conducted a survey about customers' preferences for spelling and grammar checking features in the accused products of Defendant Microsoft Corporation. Microsoft now brings a Daubert motion seeking to exclude Wecker's survey and testimony. The court agrees with Microsoft that there are significant concerns with both the structure of Wecker's survey and the way in which it was conducted. But because Ninth Circuit case law establishes that jurors are equipped to evaluate these defects themselves, the court DENIES Microsoft's motion.
See Docket No. 130.
I.
Expert testimony may only be admitted in a manner consistent with the Federal Rules of Evidence, Daubert, Kumho and more recent appellate court progeny. Federal Rule of Evidence 702 allows admission of "scientific, technical, or other specialized knowledge" by a qualified expert if it will help "the trier of fact to understand the evidence or to determine a fact in issue." When considering expert testimony, the trial court serves "as a 'gatekeeper' to exclude junk science that does not meet Federal Rule of Evidence 702's reliability standards."
Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993).
526 U.S. 137 (1999).
See Apple Inc. v. Motorola, Inc., 757 F.3d 1286 (Fed. Cir. 2014).
See Daubert, 509 U.S. at 589.
Estate of Barabin v. AstenJohnson, Inc., 740 F.3d 457, 463 (9th Cir. 2014) (citing Ellis v. Costco Wholesale Corp., 657 F.3d 970, 982 (9th Cir. 2011)). A district court's decision to admit expert testimony under Daubert in a patent case must follow the law of the regional circuit. See Micro Chem., Inc. v. Lextron, Inc., 317 F.3d 1387, 1390-91 (Fed. Cir. 2003) ("Whether proffered evidence should be admitted in a trial is a procedural issue not unique to patent law, and therefore we review the district court's decision whether to admit expert testimony under the law of the regional circuit, here the Fifth Circuit.").
An expert witness may provide opinion testimony if: (1) "the testimony is based upon sufficient facts or data;" (2) "the testimony is the product of reliable principles and methods; and" (3) "the expert has reliably applied the principles and methods to the facts of the case." "Under Daubert, the district judge is 'a gatekeeper, not a fact finder.' When an expert meets the threshold established by Rule 702 as explained in Daubert, the expert may testify and the jury decides how much weight to give that testimony." The inquiry into the admissibility of an expert opinion is a "flexible one" where shaky "but admissible evidence is to be attacked by cross examination, contrary evidence, and attention to the burden of proof, not exclusion."
Fed. R. Evid. 702; see also Sundance, Inc. v. DeMonte Fabricating Ltd., 550 F.3d 1356, 1360 (Fed. Cir. 2008) ("Patent cases, like all other cases, are governed by Rule 702. There is, of course, no basis for carving out a special rule as to experts in patent cases.").
Primiano v. Cook, 598 F.3d 558, 564-65 (9th Cir. 2010) (citing United States v. Sandoval-Mendoza, 472 F.3d 645, 654 (9th Cir. 2006)).
Id. at 564 (citing Daubert, 509 U.S. at 592-96).
A trial court thus must be sure that its review of expert testimony focuses "solely on principles and methodology, not on the conclusions that they generate." "A judge must be cautious not to overstep its gatekeeping role and weigh facts, evaluate the correctness of conclusions, impose its own preferred methodology, or judge credibility, including the credibility of one expert over another. These tasks are solely reserved for the fact finder." The Federal Circuit recently clarified that this limitation of the gatekeeping role of the judge to the exclusion of "testimony based on unreliable principles and methods is particularly essential in the context of patent damages." This is because "questions regarding which facts are most relevant or reliable to calculating a reasonable royalty are 'for the jury.'"
Daubert, 509 U.S. at 592-94, 596.
Apple Inc., 757 F.3d at 1314.
Id. at 1315.
Id. ("When the methodology is sound, and the evidence relied upon sufficiently related to the case at hand, disputes about the degree of relevance or accuracy (above this minimum threshold) may go to the testimony's weight, but not its admissibility.") (citing i4i Ltd. Partnership v. Microsoft Corp., 598 F.3d 831, 852 (Fed. Cir. 2010)).
The Ninth Circuit holds that "as long as they are conducted according to accepted principles...survey evidence should ordinarily be found sufficiently reliable under Daubert. Unlike novel scientific theories, a jury should be able to determine whether asserted technical deficiencies undermine a survey's probative value." "Treatment of surveys is a two-step process. First, is the survey admissible? That is, is there a proper foundation for admissibility, and is it relevant and conducted according to accepted principles? This threshold question may be determined by the judge. Once the survey is admitted, however, follow-on issues of methodology, survey design, reliability, the experience and reputation of the expert, critique of conclusions, and the like go to the weight of the survey rather than its admissibility."
See Southland, 108 F.3d at 1143 n.8 (internal citations omitted).
Clicks Billiards, Inc. v. Sixshooters, Inc., 251 F.3d 1252, 1263 (9th Cir. 2001) (internal citations omitted).
Sentius alleges that Microsoft infringes United States Patent Nos. RE40,731 and RE43,633 based on certain spelling and grammar check features in its products. To support Sentius' damages theory, Wecker conducted a survey that purportedly shows whether Microsoft customers would have purchased the accused products without the accused features and how much extra they are willing to pay for those features. Wecker hired ORC International, a large market research firm, to conduct his survey. ORC International conducted the survey online as part of an omnibus survey, meaning that respondents were asked questions from other surveys in addition to Wecker's survey questions.
See Docket No. 134-1 at 2-3. Sentius' damages expert Robert Mills relies on Wecker's survey to conduct his damages analysis.
See Docket No. 134-1 at 4.
See Docket No. 135-2. An omnibus survey is a survey in which respondents are asked questions from multiple different surveys.
Wecker's survey asked customers about their preferences for Microsoft's accused "'check spelling as you type' capability and 'check grammar errors as you type' capability compared to spelling and grammar check alternatives." As an alternative to the accused features, Wecker used the user-initiated spell- and grammar-checkers which Sentius' technical expert Vijay Madisetti had identified as the best non-infringing alternatives to the accused background spell and grammar checkers.
See Docket No. 134-1 at 5.
See Docket No. 134-1, Attachment D, Tables 1-4; Docket No. 145-8 at 728-31.
To determine the willingness of survey respondents to pay extra for these accused features as compared to what he believed to be the best non-infringing alternatives, Wecker asked "direct, open-ended questions." Wecker recognized that responses to direct, open-ended questions may "overstate consumers' 'true' willingness to pay." In an attempt to correct this, Wecker adjusted his willingness to pay estimates by a calibration factor.
See Docket No. 134-1 at 5.
See id.
See id at 5-6.
The survey asked respondents who had used and purchased an Office product and used the background or user-initiated spell checkers on that product to "[s]uppose the components of Microsoft Office included the user-initiated spell checker option (select Spelling and Grammar menu option), but did not include Microsoft's 'check spelling as you type' option (display a red squiggly line as you type)." The survey then asked respondents if they would have still purchased the accused products. Wecker found that approximately 14.7% of these respondents stated that they would not have purchased the accused products without the background spell checker.
See Docket 134-1 at Attachment D, Table 1, Q4.
See id.
See id.
The survey also asked respondents who had used and purchased an Office product and used the background or user-initiated grammar checkers on that product to "[s]uppose the components of Microsoft Office included the user-initiated grammar checker option (select Spelling and Grammar menu option), but did not include Microsoft's 'check grammar errors as you type' option (display green or blue squiggly line as you type)." The survey then asked respondents if they would have still purchased the accused products. Wecker found that approximately 14.8% of these respondents would not have purchased the accused products without the background spell checker. From these results, Wecker concluded that approximately 11.2% of these respondents would not have purchased the accused products if they did not include the accused features.
See id. at Attachment D, Table 3, Q8.
See id.
See id.
See id. at Attachment D, Table 3, Q4 and Q8.
II.
This court has jurisdiction under 28 U.S.C. §§ 1331 and 1338. The parties further consented to the jurisdiction of the undersigned magistrate judge under 28 U.S.C. § 636(c) and Fed. R. Civ. P. 72(a).
III.
At issue is whether Wecker's survey and opinion should be excluded from trial. The court agrees with Microsoft that there are significant concerns about both the structure of the survey and the way in which it was conducted. Because these complaints go to the weight, and not the admissibility, of the survey, the court nevertheless declines to exclude Wecker's testimony as requested.
First , the survey questions that Wecker posed to respondents were adequately tied to the subject matter that the asserted patents claim. "[T]he framing of questions for purposes of surveys is generally an issue of weight, not admissibility." At the same time, "must be outer limits to this principle." When "a description of a patent in survey...var[ies] so much from what is claimed that the survey no longer 'relate[s] to any issue in the case' and is 'not relevant and, ergo, non-helpful,'" then the survey evidence "would not 'help the trier of fact' and therefore must be excluded under Rule 702(a)." Here, the variation was not so much as to warrant exclusion.
Apple, Inc. v. Samsung Elects. Co., Ltd, Case No. 5:12-cv-00630-LHK, 2014 WL 794328, at *18 (N.D. Cal. Feb. 25, 2014) (citing Fortune Dynamic, Inc. v. Victoria's Secret Stores Brand Mgmt., Inc., 618 F.3d 1025, 1038 (9th Cir.2010)).
Id.
Id. (citing Daubert, 509 U.S. at 591); see also IP Innovation L.L.C. v. Red Hat, Inc., 705 F. Supp. 2d 687, 690 (E.D. Tex. 2010) (rejecting online consumer statements that lacked a relationship to the actual claimed technology).
Microsoft asserts that the survey questions did not distinguish between the patented and unpatented features of Microsoft's background spell and grammar check features. Sentius, Microsoft argues, does not claim that Microsoft's entire background spell and grammar check features infringe its patents, "only Microsoft's alleged use of a look-up table and its performance of certain steps in the order required by the claims." Put another way, Microsoft contends that Wecker's survey questions are overly broad because they asked respondents not about their preferences for Microsoft's spell and grammar check features "implemented in this specific way" but for Microsoft's "check spelling as you type" and "check grammar as you type" features in general.
Docket No. 130 at 8.
Id.; see also Docket No. 134-1 at 5 ("My survey directly asked respondents for their willingness to pay extra for Microsoft's 'check spelling as you type' capability and 'check grammar errors as you type' capability compared to already existing spelling and grammar check alternatives.").
The court agrees that Wecker could have more narrowly tailored his survey questions to isolate the accused aspects of Microsoft's background spell and grammar check features. But this defect is not sufficient to establish that the survey's description of the claimed invention "var[ied] so much from what is claimed that the survey no longer 'relate[s] to any issue in the case.'" Survey questions about the background spell and grammar check features are not unrelated to "any issue in the case" because, as Microsoft acknowledges, Sentius alleges that "specific aspects of the spell and grammar check features" infringe its patents.
See Apple, 2014 WL 794328, at *18 (citing Daubert, 509 U.S. at 591).
Docket No. 130 at 1; see also Apple, 2014 WL 794328 at *18 (citing Daubert, 509 U.S. at 591).
Microsoft further contends that Wecker's survey is not tied closely enough to Sentius' patented technology because it did not ask questions about non-infringing alternatives that could implement the accused aspects of the background spell and grammar check features. However, in Apple v. Samsung, the court rejected the defendant's argument that plaintiff's survey was improper in part because the record allowed the plaintiff to argue that it had probed customers about "the relevant choice" between the accused feature and the best non-infringing alternative. Sentius' contention that it has identified the relevant choice for consumers is not as strong as that of the plaintiff in Apple v. Samsung because there the defendant "presented no evidence" to rebut the plaintiff's assertion that the defendant could not have implemented the proposed non-infringing alternative. In contrast, Microsoft has offered evidence that it could have implemented "several" non-infringing alternatives. But Sentius is not required to prove that it has presented consumers with the correct choice in order to survive a Daubert motion, only that it can "argue" that its selected choice is correct. Microsoft's "dissatisfaction with the description of the patented features in the survey...goes to weight, not admissibility."
See Docket No. 130 at 7-11.
See Apple, 2014 WL 794328, at *18-19. The defendant argued that the plaintiff's survey had "failed to capture an essential limitation of the asserted claim." Id. at *18. The asserted claim required three background synchronization components but plaintiff's survey asked respondents to choose between a device with a single background synchronization component and a device with no background synchronization feature. See id. at *18-19. Because the unrebutted evidence showed that defendant "could not have implemented a single-component non-infringing alternative" the court found that "[plaintiff] can argue that, for [defendant's] consumers, the technical limitations of [defendant's] devices rendered the relevant choice to be between a device with background synchronization and a device without background synchronization (not a choice between three-component background synchronization or one-component background synchronization)." Id. at *19. The court found that the survey expert's description of the claim at issue was "not improper." Id.
See Apple, 2014 WL 794328, at *19.
See Docket No. 130 at 7-8; Docket No. 131-10 at paras. 656-668.
See Apple, 2014 WL 794328, at *19.
See id. at *18 (internal citations omitted). Microsoft also seeks to exclude Wecker's survey under Fed. R. 403. See Docket No. 130 at 11-11. Under Rule 403, a "court may exclude relevant evidence if its probative value is substantially outweighed by a danger of one or more of the following: unfair prejudice, confusing the issues, misleading the jury, undue delay, wasting time, or needlessly presenting cumulative evidence." Fed. R. Evid. 403. The court will not exclude the survey under Rule 403 because it holds that "discrepancies between the scope of [Sentius'] patent claims and the survey questions" are not "so confusing to the jury as to substantially outweigh the survey's probative value." See Apple, 2014 WL 794328, at *18.
Second , Wecker's survey is based on reliable methodologies. Microsoft contends that Wecker's survey must be excluded because it asked respondents only about their preferences for the two accused features when the accused products include thousands of features, was part of an omnibus survey and lacked quality control measures. As with Microsoft's complaint that Wecker's survey questions did not accurately capture the value of the claimed inventions, these allegations have merit. But because "the Ninth Circuit has held, on multiple occasions, that '[c]hallenges to survey methodology,' including 'the format of the questions or the manner in which it was taken, bear on the weight of the evidence, not its admissibility'" the court will not exclude Wecker's survey on these grounds.
See Docket No. 130, at 12-16.
See Wells Fargo & Co. v. ABD Ins. & Fin. Servs., Inc., Case No. 4:12-3856-PJH, 2014 WL 4312021, at *3 (N.D. Cal. Aug. 28, 2014) (citing Fortune, 618 F.3d at 1036; Wendt v. Host Int'l, Inc., 125 F.3d 806, 814 (9th Cir. 1997)).
Microsoft argues that Wecker's survey is flawed because it asked respondents about their preferences for the accused spell and grammar check features even though the accused Microsoft Office applications have "thousands of features." Microsoft says that this type of survey—which Wecker called the "direct approach" and the "direct method"—does not pass muster under Daubert because it is not the "generally accepted methodology" and results in "figures that are unreliably biased to favor Sentius." But Wecker cited to a 2011 article that described the direct method as one of several "commonly used approaches" for measuring willingness to pay. Microsoft takes issue with the sufficiency of this support, but cites to no case that establishes that a direct method survey is not a generally accepted methodology to analyze respondents' willingness to pay. Microsoft also claims that the generally accepted practice is to use a conjoint analysis but cites to no case in which a court has excluded a survey because an expert should have conducted a conjoint survey instead of a direct method survey. Microsoft further asserts that Sentius has not shown that real world commercial decisions have turned on the type of survey Wecker used but offers no authority to establish that Sentius has the duty to show this in order to survive a Daubert motion.
See Docket No. 130 at 12.
See Docket No. 134-1 at 5. Surveys that use the direct approach ask "consumers directly to state their [willingness to pay] for a specific product through, for example, an open-ended (OE) question format." Docket No. 146-9 at 173.
Docket No. 130 at 13; see also Daubert, 509 U.S. at 594 ("Widespread acceptance can be an important factor in ruling particular evidence admissible, and 'a known technique which has been able to attract only minimal support within the community,' may properly be viewed with skepticism.") (internal citations omitted); Liquid Dynamics Corp. v. Vaughan Co., Inc., 449 F.3d 1209, 1220-21 (Fed. Cir. 2006) (identifying a methodology's general acceptance as a Daubert factor).
See Docket No. 134-1 at 5 n.8; Docket No. 146-9 at 173. Wecker also cited to evidence that regulatory agencies measures damages and scale restoration by using "hypothetical questions posed by surveys to ascertain how much respondents would be willing to pay to preserve the natural resources." See Docket No. 134-1 at 5 n.8; Docket No. 146-10 at 2.
See Docket No. 130 at 12-13; Docket No. 162 at 10-11.
See Docket No. 130 at 12-13; Docket 135-5 at 2-3:
Conjoint (trade-off) analysis is one of the most widely-used quantitative methods in Marketing Research. It is used to measure preferences for product features, to learn how changes to price affect demand for products or service, and to forecast the likely acceptance of a product if brought to market.
Rather than directly ask survey respondents what they prefer in a product, or what attributes they find most important, conjoint analysis employs the more realistic context of respondents evaluating potential product profiles.
See Docket No. 130 at 12-13; Docket No. 145-7 at 14.
See Docket No. 130 at 12-13; Docket No. 162 at 10-11.
Microsoft's allegations that Wecker improperly used an omnibus survey and lacked adequate quality control measures likewise are not reasons to exclude Wecker's survey under Daubert. As with Wecker's use of a direct method survey, Microsoft cites to no case that has excluded surveys on these grounds. The court agrees that these factors likely impacted the quality of Wecker's survey, but these concerns go to the weight, not admissibility of the survey.
See Docket No. 130 at 14-16.
See Droplets, Inc. v. Overstock.com, Inc., Case No. 2:11-cv-00401-JRG-RSP, Docket No. 341, at *4 (E.D. Tex. Jan. 7, 2015) ("The fact that Dr. Schwartz's survey may be imperfect or conducted in an alternative fashion does not automatically support Defendants' conclusion that Dr. Schwartz's methodology was unreliable.").
Microsoft also claims that Wecker's survey is defective because it presents risk of "hypothetical bias" because it used the direct approach rather than a conjoint analysis. This hypothetical bias, Microsoft claims, may have caused respondents to focus unduly on the accused features, resulting in figures that are "unreliably biased to favor Sentius." But Microsoft cites to no case in which a court excluded a survey because of concerns with hypothetical bias. In fact, courts have declined to exclude surveys on the grounds that hypothetical bias may be present. The effect of hypothetical bias on Wecker's survey is the proper subject of a cross-examination, not a reason to exclude Wecker's survey under Daubert.
See Docket No. 130 at 1-2. Microsoft says that "hypothetical bias" is the notion that "people asked to focus on one particular feature in a product will naturally overstate the importance of that feature." Id. at 2.
See Docket No. 130 at 1-2, 12-13.
See, e.g., TV Interactive Data Corp. v. Sony Corp., 929 F. Supp. 2d 1006, 1025-26 (N.D. Cal. 2013) (denying motion to exclude expert's survey where defendant argued that "the exclusion of the majority of product attributes...led respondents to overvalue each of the product attributes that were tested" for a willingness to pay survey because the dispute about whether respondents had overvalued the product features was a "'battle of the experts' for the jury to decide"); Hartle v. FirstEnergy Generation Corp, Case Nos. 08-1019, 08-1025, 08-1030, 2014 WL 1317702, at *6 (W.D. Pa. Mar. 31, 2014) reconsideration denied sub nom Patrick v. FirstEnergy Generation Corp, Case Nos. 08-1025, 08-130, 2014 WL 5463885 (W.D. Pa. Oct. 27, 2014) (finding that defendants' arguments regarding hypothetical bias went only to the weight, and not the admissibility of a willingness to pay survey used to quantify harm from air pollution).
Third , Wecker's use of a "calibration" factor to adjust the hypothetical bias presented in his survey was reliable. Wecker recognizes that respondents may overestimate their willingness to pay when answering open-ended questions about a particular feature of a product. Wecker's survey presents this risk because it asked open-ended questions about respondents' willingness to pay for spell and grammar check features. Wecker attempted to fix this by dividing his survey results by a calibration factor of 1.46, a figure he calculated based on results of a meta-analysis of twenty-nine experimental studies. Microsoft claims that Wecker's use of a calibration factor is unreliable because Wecker did not conduct a study to determine the value of the factor and because the articles upon which Wecker based his "calibration approach" do not show that this approach is generally accepted in the survey field. However, as stated above, Microsoft has cited to no case establishing that the presence of hypothetical basis is sufficient to exclude a survey under Daubert. So the way in which an expert adjusts the degree to which hypothetical bias caused respondents to overstate their willingness to pay is not a proper reason for exclusion of the survey either. Although the court agrees with Microsoft that there are reasons to conclude that Wecker's use of a calibration factor may be flawed, it will not exclude the survey on this basis.
See Docket 134-1 at 5.
See id.
See id. at 5-6.
See Docket No. 130 at 16-19.
--------
IV.
Because Microsoft's complaints about Wecker's survey go to weight, not admissibility, the court DENIES Microsoft's motion to exclude Wecker's survey and testimony. SO ORDERED. Dated: January 23, 2015
/s/_________
PAUL S. GREWAL
United States Magistrate Judge