“Vigorous cross-examination, 4 || presentation of contrary evidence, and careful instruction on the burden of proof are the traditional 5 and appropriate means of attacking shaky but admissible evidence.”
How later courts described this case
- “Vigorous cross-examination, 4 || presentation of contrary evidence, and careful instruction on the burden of proof are the traditional 5 and appropriate means of attacking shaky but admissible evidence.”
- at the certification stage, it is 3 15 || not necessary to show that the expert’s method for determining class damages will work with 16 || certainty
Written by the judges who cited it.
The opinion
1
2
3 UNITED STATES DISTRICT COURT
4 NORTHERN DISTRICT OF CALIFORNIA
5 SAN JOSE DIVISION
6
7 RENE CABRERA, ET AL., Case No. 5:11-cv-01263-EJD
8 Plaintiffs, ORDER RE GOOGLE’S MOTIONS TO
STRIKE REPORTS OF SAUL
9 v. SOLOMON
10 GOOGLE LLC,
Re: Dkt. Nos. 442, 596
Defendant.
11
12 Defendant Google LLC (“Google”) has filed two separate motions to strike the expert reports
13 of Saul Solomon pursuant to Federal Rule of Evidence 702 and Daubert v. Merrell Dow
14 Pharmaceuticals, 509 U.S. 579 (1993). ECF 442, 596.1 Plaintiffs oppose the motions (ECF 455,
15 609). For the reasons stated below, the Court grants Google’s motions in part.
16 I. BACKGROUND
17 Because the parties are intimately familiar with the history of this case, this Order sets
18 forth only what is necessary to resolve the issues presented in Google’s motion.
19 Saul Solomon (“Solomon”) is Plaintiffs’ damages expert. He issued reports on July 10,
20 2015 (ECF 441-10), January 18, 2018 (ECF 424-80), May 15, 2018 (ECF 320-14), and November
21 8, 2018 (the “Cabrera Data Report”) (ECF 424-86). Google’s Mot. at 4. In the July 10, 2015
22 report, Solomon states his opinion that restitution and contract damages resulting from Google’s
23 alleged misconduct relating to Smart Pricing and Location Targeting are measurable and can be
24 reasonably and reliably determined. ECF 441-10 at 3. Further, he states that damages and
25
26
1 Google’s motion filed as ECF 442 challenges both Solomon’s opinions and the opinions of
27 another one Plaintiffs’ experts, Dr. James L. Gibson (“Gibson”). The Court will address the
portion of Google’s motion challenging Gibson’s opinions in a separate order.
1 restitution can be computed for all putative Class Members by applying a uniform methodology
2 that relies solely on Google’s data and requires no individual inquiry of any putative Class
3 Member. Id. Solomon opines that restitution for the Location Targeting claim can be measured in
4 two ways. First, Solomon says he can apply the “partial refund model” by summing the amount
5 paid by Class Members for all Out-of-Area Clicks. Id. Second, Solomon says he can estimate
6 restitution based upon the “net profits model” by applying an estimate of the applicable
7 incremental profit margin actually realized by Google on advertising revenue to the amount paid
8 by Class Members for all Out-of-Area Clicks. Id. Solomon also explains his methodology for
9 computing restitution for the Smart Pricing claim. Id. at 3-4.
10 In his January 18, 2018 report, Solomon opines in pertinent part that the revised class
11 definitions do not impact his earlier opinions. ECF 424-80. He also proposes a third method to
12 calculate restitution for the Location Targeting claim, the Smart Pricing-based restitution model
13 (“SPR Model”), which measures the difference between (i) the amount Class Members actually
14 paid for Out-of-Area Clicks and the amount Google’s Smart Pricing algorithms estimate that a
15 rational advertiser might have been willing to pay for the mistargeted clicks had the advertiser
16 been forced to pay for those Out-of-Area Clicks. Id. at 4, 9-10. The SPR Model “assumes that
17 advertisers should have to pay for the mistargeted clicks accrued due to Google’s improper
18 conduct and despite their express ‘0’ bid, and reprices the ads that were placed outside the areas
19 selected by advertisers using Google’s Smart Priced data.” Id. at 5. Solomon states that by using
20 Google’s data, as compiled by Dr. Gibson, he can reasonably and reliably measure the amount of
21 restitution under the SPR Model for all putative Class Members resulting from any Out-of-Area
22 Clicks. Id. at 8.
23 In his May 15, 2018 report, Solomon asserts that the analyses and key assumptions
24 underlying the opinions of Google’s expert, Dr. Lawrence Wu, are flawed and speculative. ECF
25 320-14 at 3-5. Solomon also takes issue with the analysis and conclusion of Google’s expert, Dr.
26 Paul Milgrom. Id. at 5.
27 In the Cabrera Data Report, Solomon measures the damages and restitution for Plaintiff
1 Rene Cabrera (“Cabrera”) resulting from Google’s alleged misconduct using the Class Period
2 click data Google produced (the “Cabrera Clicks”). ECF 424-86. Pertinent to the Location
3 Targeting claim, Solomon applies the partial refund model and concludes that Cabrera’s restitution
4 for Out-of-Area Clicks during the Class Period totals $88.73. Id. at 4. Applying the net profits
5 model leads to a total of $62.66 in restitution for Cabrera. Id. Applying the SPR Model, Solomon
6 concludes that Cabrera’s restitution totals $8.69. Id. Solomon also provides a $8.90 damages
7 calculation for Cabrera’s Smart Pricing claim. Id. at 5.
8 After Solomon issued the Cabrera Data Report, Plaintiffs filed the Fifth Amended
9 Complaint naming RM Cabrera Company, Inc. (“RMC”) f/k/a Training Options, Inc. as an
10 additional plaintiff. In his November 18, 2021 Supplemental Report (ECF 579-3), Solomon
11 expresses his understanding, based on Cabrera’s deposition testimony, that all the Cabrera Clicks
12 were purchased by RMC for RMC’s business.” ECF 579-3 ¶ 3. He then expresses his opinion
13 that the “measurement of damages and restitution” presented in his earlier reports regarding
14 Cabrera is equally applicable to RMC. Id.
15 II. STANDARDS
16 Federal Rule of Evidence 702 permits opinion testimony by an expert if the proponent
17 demonstrates that the expert is qualified and (a) the expert’s scientific, technical, or other
18 specialized knowledge will help the trier of fact to understand the evidence or to determine a fact
19 in issue; (b) the testimony is based on sufficient facts or data; (c) the testimony is the product of
20 reliable principles and methods; and (d) the expert has reliably applied the principles and methods
21 to the facts of the case. Fed. R. Evid. 702. An expert witness may be qualified by “knowledge,
22 skill, experience, training, or education.” Id. The proponent of expert testimony has the burden of
23 proving admissibility in accordance with Rule 702. Fed. R. Evid. 702 advisory committee’s note
24 to 2000 amendment.
25 Rule 702 “clearly contemplates some degree of regulation of the subjects and theories
26 about which an expert may testify.” Daubert, 509 U.S. at 589–90. Under Daubert, the Court
27 exercises a gatekeeping function to ensure an expert’s proffered testimony is relevant and reliable.
1 United States v. Valencia-Lopez, 971 F.3d 891, 897–98 (9th Cir. 2020). Rule 702 and Daubert,
2 however, are not “guarantees of correctness.” i4i Ltd. P’ship v. Microsoft Corp., 598 F.3d 831,
3 855 (Fed. Cir. 2010). “[T]he case law—particularly Ninth Circuit case law—emphasizes that a
4 trial judge should not exclude an expert opinion merely because he thinks it’s shaky, or because he
5 thinks the jury will have cause to question the expert’s credibility. So long as an opinion is
6 premised on reliable scientific principles, it should not be excluded by the trial judge.” In re
7 Roundup Prods. Liab. Litig., 390 F. Supp. 3d 1102, 1109 (N.D. Cal. 2018).
8 III. DISCUSSION
9 Google challenges Solomon’s damages calculations for both the Smart Pricing claim and
10 the Location Targeting claim. Google argues that Solomon’s opinions are neither admissible nor
11 probative expert testimony for several reasons. First, Google argues that Solomon’s opinions
12 regarding Smart Pricing damages rest on a fundamental misconception of how Smart Pricing
13 works. Second, Google contends that Solomon’s opinions regarding Location Targeting damages
14 are based on the false premise that no advertiser has ever benefited from Out-of-Area Clicks.
15 Third, Google contends that Solomon’s opinions regarding class damages are invalid because they
16 are based on a single potential class member, Cabrera, which Google contends is insufficient as a
17 matter of law to show that damages may be proved with common class-wide evidence.
18 A. Solomon’s Opinions re Smart-Pricing Claim
19 Google’s challenges to Solomon’s opinions regarding Smart Pricing damages depend in
20 large part on the strength of the legal arguments in its motion for summary judgment and in
21 opposition to Plaintiffs’ motion for class certification. Therefore, the Court will defer ruling on
22 Google’s challenges to Solomon’s opinions regarding Smart Pricing damages until the Court
23 addresses the parties’ respective motions.
24 B. Solomon’s Opinions re Location Targeting Claim
25 In calculating restitution under the UCL, California law “requires only that some
26 reasonable basis of computation of damages be used, and the damages may be computed even if
27 the result reached is an approximation.” Pulaski & Middleman, LLC v. Google, Inc., 802 F.3d
1 979, 989 (9th Cir. 2015), cert. denied, 136 S. Ct. 2410 (2016) (“Pulaski”) (quoting Marsu, B.V. v.
2 Walt Disney Co., 185 F.3d 932, 938-39 (9th Cir. 1999)). Solomon’s computations satisfy this
3 standard.
4 Contrary to Google’s assertion, Solomon’s opinions regarding damages for the Location
5 Targeting claim are based on more than just “intuition” and “logic.” Solomon explains in his May
6 15, 2018 declaration:
7 The basis for my conclusion that Out-of-Area Clicks have zero
value to advertisers is not my “intuition” and “logic” as Dr. Wu
8 erroneously asserts, but rather the value is evidenced by advertisers’
explicit selection to exclude these clicks from [their] ads indicating
9 they chose to pay nothing for them and Google’s policy of providing
a full refund of the click cost for Out-of-Area Clicks to advertisers
10 (regardless of whether the click resulted in a conversion).
11 ECF 320-14 at 4. Solomon explains further in his January 14, 2018 declaration:
12 When given the option to specify the geographic location(s) in
which they want their ads to appear for Location Targeting,
13 advertisers will effectively exclude the locations they do not select.
It is through this selection of geographic location(s) in which they
14 want their ads to appear that the advertisers express to Google that
they have chosen to spend $0 advertising monies on any Out-of-
15 Area Clicks. Thus, the value of $0 is evidenced and assigned to Out-
of-Area Clicks first and foremost by the advertisers themselves. I
16 have relied upon the advertisers’ geographic location selections to
indicate the value determined for Out-of-Area Clicks by the
17 advertisers themselves. The logic involved here is merely the
assertion that, by deciding to pay zero for Out-of-Area Clicks, the
18 advertisers have determined that the benefit they expected to receive
for these clicks does not exceed the cost, resulting in a value of zero.
19
20 June 14, 2018 Decl. of Saul Solomon in Response to Google’s Mot. to Strike Solomon Expert
21 Reports, ECF 454-10, ¶ 14. That Google gave a refund to one advertiser who complained about
22 Out-of-Area clicks (ECF 454-10 at 9-10) also tends to support Solomon’s opinions, even if only
23 slightly.
24 Further, contrary to Google’s assertion, Solomon never “admits” that “it’s probably not
25 true that advertisers value out-of-area clicks at zero.” Mot. at 21. Instead, Solomon testifies
26 during his deposition that: “If you’re asking me is it conceivable that an out-of-area click could
27 have provided some value to an advertiser, I would say it’s possible after the fact. 3/14/2018
1 Solomon Dep., ECF 441-4, at 48:22-25. This concession does not render Solomon’s opinions so
2 || unreliable that they are subject to exclusion. Google’s remedy is to cross-examine Solomon and
3 put forth its own rebuttal evidence. See Daubert, 509 U.S. at 596 (“Vigorous cross-examination,
4 || presentation of contrary evidence, and careful instruction on the burden of proof are the traditional
5 and appropriate means of attacking shaky but admissible evidence.”).
6 1. Sample Size
7 Google contends that Solomon’s opinions are unreliable for purposes of calculating
8 damages to Class Members because they are based on an analysis of account information from just
9 Cabrera and former Plaintiff Rick Woods. This is inaccurate. Solomon relies on, but does not
10 || discuss in significant detail, the opinions offered by Dr. Gibson, who analyzed Google’s Data
11 Sample. ECF 320-14, Ex. A.
12 Google also complains that Solomon has not actually performed a full calculation of class-
5 13 wide damages, but that is not required at this time. Spann v. J.C. Penney Corp., 307 F.R.D. 508,
14 529 (C.D. Cal. 2015), modified, 314 F.R.D. 312 (C.D. Cal. 2016) (at the certification stage, it is
3 15 || not necessary to show that the expert’s method for determining class damages will work with
16 || certainty). It is sufficient that Solomon has set forth likely methods for calculating class damages
3 17 for the Location Targeting claim. Jd.
18 |l Iv. CONCLUSION
19 For the reasons stated above, Google’s motions are DENIED to the extent they challenge
20 || Solomon’s opinions regarding Location Targeting damages. Google’s motions are DEFERRED
21 to the extent its motions challenge Solomon’s opinions regarding Smart Pricing damages.
22 IT IS SO ORDERED.
23 Dated: September 26, 2022
24
25 aD.
EDWARD J. DAVILA
26 United States District Judge
27
28 || Case No.: 5:11-cv-01263-EJD
ORDER RE GOOGLE’S MOTIONS TO STRIKE REPORTS OF SAUL SOLOMON